Hughes et al. BMC Research Notes 2010, 3:46 http://www.biomedcentral.com/1756-0500/3/46
SHORT REPORT Open Access Clinical experience, infection control practices and diagnostic algorithms for poxvirus infections - an Emerging Infections Network survey Christine M Hughes1*, Edith R Lederman1,2, Mary G Reynolds1, Inger K Damon1, R Ryan Lash1, Susan E Beekmann3, Philip M Polgreen3, the Infectious Diseases Society of America’s Emerging Infections Network
Abstract Background: In order to determine how best to tailor outreach messages about poxvirus diagnosis and infection control for health practitioners, we surveyed infectious disease physicians in the Infectious Diseases Society of America’s Emerging Infections Network. Findings: Surveys consisting of two unknown case scenarios designed to raise suspicion for monkeypox and orf were distributed to the 1,080 members of the EIN. The surveys contained questions pertaining to which diagnostic tests, points of contact, and transmission precautions they would likely utilize during patient evaluation. Basic response rates and frequencies of responses were calculated. Comparisons of the survey responses were made using the chi-square test. Of the 212 members who responded (20% response rate), significantly more respondents indicated that they would request diagnostic testing in the context of the monkeypox case scenario as compared to the orf case scenario. A significantly higher number of respondents indicated they would institute droplet or airborne precautions for the monkeypox case as opposed to the orf case scenario. Conclusions: This survey provided an opportunity for public health practitioners to gain insight into physician approaches to evaluation, diagnosis and reporting of suspected poxvirus-associated infections. This survey identified key areas in which public health practitioners can better serve physicians by focusing on education. As a result we were able to identify potential knowledge gaps and deficits in the availability of useful resources to facilitate accurate case identification and management.
Findings but may also occur in larger communities with live ani- Inthewakeofthe2003U.S.monkeypoxoutbreakand mal markets, petting zoos and small-scale animal hus- renewed concerns regarding bioterrorism, poxvirus bandry [1]. Inadvertent Vaccinia virus infections also infections have garnered increased attention from medi- occur in the United States. Vaccinia is the primary com- cal and public health professionals alike. ponent of the smallpox vaccine and infections can occur There are multiple poxviruses of significance to following contact with a recent vaccinee or via exposure human health that occur in the United States. These tothevirusinalaboratory[2-4]. Additionally, the cur- include Molluscum contagiosum virus, which causes rent oral rabies vaccine (ORV), used to prevent the common viral infections of the skin, and various para- spread of terrestrial rabies along the Eastern seaboard, poxviruses, such as Orf virus and Pseudocowpox virus, consists of a recombinant vaccinia virus containing the which are zoonotic entities associated with domestic rabies virus glycoprotein gene. Two human infections ruminants (e.g., sheep, goats, cattle). Human parapox- following contact with ORV has been reported [5-7]. virus infections occur primarily in rural communities, Importation of poxviruses from abroad is also a con- cern. This occurred in 2003 when monkeypox infected * Correspondence: [email protected] African rodents were imported to the United States 1 Centers for Disease Control and Prevention, National Center for Zoonotic, resulting in a outbreak of monkeypox [8,9]. Monkeypox Vector-borne, and Enteric Diseases, Division of Viral and Rickettsial Diseases, 1600 Clifton Road, NE Atlanta, Georgia, 30333, USA virus is a communicable orthopoxvirus which can cause
© 2010 Hughes et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Hughes et al. BMC Research Notes 2010, 3:46 Page 2 of 6 http://www.biomedcentral.com/1756-0500/3/46
systemic infections in humans similar to Variola virus During February and March of 2007, surveys were dis- (smallpox). This event marked the first time that human tributed by e-mail and facsimile to the 1,080 EIN. Parti- monkeypox infections had been observed in the Wes- cipants were encouraged to use any reference material tern Hemisphere. There have also been two reports of they deemed necessary to complete the query and were Tanapox virus (a yatapoxvirus) infections in travelers allowedtologinandoutofthesurveyiftheyrequired returning to the U.S. from Africa [10,11] more time. Many poxvirus infections share common clinical fea- Our survey was developed to assess clinical experi- tures (e.g., vesiculo-pustular or nodular rash lesion char- ence, infection control practices, and diagnostic algo- acteristics) but have differing risks for person-to-person rithms related to poxviruses.Thesurveyconsistedof transmission, thus necessitating different infection con- two unknown case scenarios designed to raise suspicion trol measures. Poxvirus infections can be confused with for monkeypox and orf (Figure 1A and 1B, respectively) other infections or conditions. This underscores the with corresponding questions pertaining to which diag- importance of laboratory diagnostic evaluation when nostic tests and transmission precautions they would poxvirus infections are suspected. Diagnostic testing for likely use during patient evaluation [Additional file 1: poxvirus infections is available; however most are only Copy of survey]. Members were queried as to their likely available at specialized reference centers (e.g., Labora- immediate points of contact for reporting of the suspi- tory Response Network facilities, Centers for Disease cious illnesses. In order to gain some insight on the Control and Prevention (CDC)). poxvirus experience for physicians in various regions of We surveyed infectious disease physicians in the Infec- the country, members were also asked to document the tiousDiseasesSocietyofAmerica’s(IDSA)Emerging types of poxviruses they have ever seen in their practice. Infections Network (EIN) to gather insight on current Since the first distribution of the survey resulted in a diagnostic and infection control practices for suspect low response rate, respondents were given the option of poxviruses in order to tailor outreach messages to health omitting their name from the second round of distribu- practitioners. In addition, we sought to determine what tion. They were also given a choice of submitting their resources were readily available to physicians to assist in state and practice type in lieu of their name. Therefore, evaluation of suspected poxvirus infections. This survey since demographic data is linked by name, some respon- also allowed us to gain some insight into the frequency dents do not have linked demographic data. and spectrum of poxvirus infections seen by this net- Basic response rates for demographic variables and work of consultants. frequencies of specific responses for each survey ques- IDSA’s EIN is a provider-based emerging infections tion were calculated. Denominators varied for several sentinel network of adult and pediatric infectious disease questions as members did not always respond to all the consultants. This network was established through a survey questions. Comparisons of responders and non- Cooperative Agreement Program Award in 1995 from responders, as well as comparisons in responses to the the CDC [12]. two case scenarios were made using the chi-square test.
Figure 1 Case scenario pictures. A) Monkeypox case scenario: 23 year old male medical student with several pustular skin lesions (upper and lower extremities including volar surfaces), lymphadenopathy, fever, chills, backache, malaise; he recently returned from Democratic Republic of Congo where he examined patients with undiagnosed febrile rash illness. Photo by Dr. Janet A. Fairley, 2003. B) Orf case scenario: 42 year old male with two large nonpruritic, painless vesicular lesions on thumb and forefinger; he denies other symptoms, works on farm and recently purchased juvenile goats at auction (some of which had ulcers on their oral mucosa). Photo by Dr. Susan Meidl, 2006 Hughes et al. BMC Research Notes 2010, 3:46 Page 3 of 6 http://www.biomedcentral.com/1756-0500/3/46
A p-value of less than 0.05 was considered statistically Table 1 Geographic and practice characteristics of significant. poxvirus survey respondents vs. entire EIN participant Of the 1,080 EIN members to whom surveys were dis- base Variable Respondents Total EIN Response tributed, 212 (20%) returned completed surveys. Of † these 212 surveys, 29 (13.7%) of them were returned (n = 212), no. (n = 1076 ), no. rate (%) (%) without a name and were not able to be linked to corre- Type of practice sponding demographic data. Respondents included phy- Adult 141 (77.5%) 786 (73.1%) 17.94% sicians from all nine U.S. Bureau of Census divisions, Pediatric 34 (18.7%) 213 (19.8%) 15.96% along with two respondents from Canada (Table 1). Adult & Pediatric 7 (3.9%) 75 (6.97%) 9.33% Those with less than 10 years of experience were signifi- Other 0 2 (0.18%) cantly less likely to respond to the survey compared to those with over 10 years experience (p = 0.02). Those Practice Location who teach were significantly more likely to respond to Rural 11 (7.5%) 48 (6.8%) 22.92% the survey then those who don’t (p = 0.004). The lowest Suburban 40 (27.4%) 150 (21.3%) 26.67% response rates came from those in an urban setting, Urban 93 (63.7%) 496 (70.6%) 18.75% those who do not teach, and those with less than 10 combination 2 (1.4%) 9 (1.3%) 22.22% years of experience. EIN members from the New Eng- land and Mid Atlantic region were the least likely to Teach respond to the survey, while those from the East North Yes 131 (72.8%) 637 (61.6%) 20.57% Central, West South Central, and the Mountain region No 49 (27.2%) 397 (38.4%) 12.34%* were the most likely to respond. Of the 212 respondents, significantly more, 22%, Practice Type would rely on clinical diagnosis alone for etiologic deter- Academic 105 (52.2%) 404 (55.9%) 25.99% mination in the context of the orf case scenario in con- Private 84 (41.8%) 264 (36.7%) 31.82% trast to the monkeypox case scenario, 3% (p < 0.0001). Other 12 (6.0%) 54 (7.5%) 22.22% However, the majority of respondents indicated that they would likely request laboratory testing for determi- Region nation of poxvirus etiology for both case scenarios. Sig- New England 13 (6.3%) 92 (8.6%) 13.83% nificantly more respondents indicated that they would Mid Atlantic 28 (13.5%) 196 (18.2%) 14.29% likely request polymerase chain reaction (PCR) based East North Central 36 (17.3%) 144 (13.4%) 25.00% diagnostic technologies in the context of the monkeypox West North 16 (7.7%) 75 (7.0%) 21.33% case scenario as compared to the orf case scenario (87%, Central vs.67.9%) (p < 0.001) (Table 2). Significantly more South Atlantic 34 (16.4%) 214 (19.9%) 15.89% respondents, 66.0%, indicated that they would choose a East South Central 12 (5.8%) 49 (4.5%) 24.49% state or federal lab in the monkeypox case for PCR test- West South 18 (8.7%) 72 (6.7%) 25.00% ing, versus 34.0% for the orf case (p < 0.001). Central Significantly more respondents would order a serologi- Mountain 14 (6.7%) 54 (5.0%) 25.93% cal test for the monkeypox case scenario (72.6%) than Pacific 35 (16.8%) 160 (14.9%) 21.88% for the orf case (38.7%) (p < 0.0001). The majority indi- Canada 2 (1%) 13 (1.2%) 15.38% cated that they would likely utilize a state or federal lab Puerto Rico 0 (0%) 6 (0.6%) for serologic testing in the context of the suspected monkeypox case (50%) and the suspected orf case No. yrs practice (21.7%). A significantly higher number of respondents <10 yrs 9 (8.0%) 74 (16.9%) 12.16%* would also order a culture or histopathology for the 10-20 yrs 51 (45.5%) 162 (37.0%) 31.48% monkeypox case (73.6%) in contrast to the orf case 21-30 yrs 38 (33.9%) 147 (33.6%) 25.85% (43.9%) (p < 0.0001). An in-house or local academic lab 31+ yrs 14 (12.5%) 55 (12.6%) 25.45% wasmostlikelytobepickedforbothcasescenarios Note: # of respondents does not equal 21 for some variables due to missing (46.7% and 31.6% respectively). information † When asked what type of precautionary measures they Demographic data was available for 1076 of the 1080 EIN participants ’ * Variable group has significantly lower response rate compared to the rest of would likely institute during examination of the patient s the responses for that variable combined Hughes et al. BMC Research Notes 2010, 3:46 Page 4 of 6 http://www.biomedcentral.com/1756-0500/3/46
Table 2 Diagnostic ordering preferences for the two case scenarios in the EIN poxvirus survey Monkeypox Scenario Orf Scenario Diagnostic Test Lab utilized # %* # %* PCR In-house/local academic institution 61 28.8% 70 33.0% State/Federal 140 66.0% 72 34.0% Commercial reference lab 25 11.8% 26 12.3% In-house/local academic institution 37 17.5% 17 8.0% Serology State/Federal 106 50.0% 46 21.7% Commercial reference lab 36 32.1% 28 13.2% In-house/local academic institution 99 46.7% 67 31.6% Culture/Histopathology State/Federal 68 32.1% 23 10.8% Commercial reference lab 8 3.8% 6 2.8% Note: Monkeypox specific PCR, culture and histopathology are currently available at CDC. Most state reference laboratories are able to perform orthopoxvirus generic PCR. Serological testing at CDC is orthopoxvirus generic. Orf specific PCR is currently available at CDC while orf serology is available at an outside lab (Viral and Rickettsial Disease Laboratory, California Department of Public Health, Richmond, CA). * Percent of total responders. Numbers do not add up to 100% as respondents were able to pick multiple choices. described in hypothetical scenario A or B, a significantly When respondents were asked whom they might report higher proportion of respondents indicated that they initial suspicion of poxvirus-associated illness, a signifi- would choose routine only precautions for the orf case cantly higher proportion of respondents indicated that scenario (23.7%), as opposed to the monkeypox case they would report the suspected monkeypox case (88.2%) scenario (8.0%) (p < 0.0001) (Figure 2). A significantly as opposed to the orf case (66.8%) (p < 0.0001), with most higher number of respondents indicated they would respondents choose the state or local health department as institute droplet or airborne precautions for the mon- the first point of contact for either scenario. keypox case (22.6% & 48.1% respectively), as opposed to Respondents were also asked about the relative fre- the orf case scenario (6.6% & 7.1%) (p < 0.0001). quency with which they have ever encountered different
Figure 2 Likely precautionary measures indicated for each hypothetical case scenario. The numbers in each portion of the diagram represent the number of respondents choosing that combination of precautions. The appropriate choices have been bolded. CDC recommends a combination of standard, contact, and droplet precautions for possible monkeypox virus (or other systemic orthopoxvirus) infections. In addition, because of the theoretical risk of airborne transmission, Airborne Precautions should be applied whenever possible. CDC recommends standard precautions for possible orf virus infections. Hughes et al. BMC Research Notes 2010, 3:46 Page 5 of 6 http://www.biomedcentral.com/1756-0500/3/46
poxvirus-associated illnesses in their practices. The major- and different professional groups had different response ity (96.7%) had seen at least one case of molluscum conta- rates. Those who did not teach and those with less then giosum with 74.5% having seen five or more cases. 6% 10 years of experience had a significantly lower response reported having seen at least 1 case of monkeypox, with rate. Infectious disease physicians with less experience six of these physicians practicing in the Midwest region are probably less likely to have encountered some of (Figure 3). 4.7% reported having seen a case of vaccinia in these uncommon diseases and thereby are less likely to a lab worker while 8.9% had seen a vaccinia infection in a see the relevance of this query. Response rates also var- social contact of a vaccinee. Respondents from each of the ied between regions, which could be due to differences EIN regions reported having seen orf, with the largest in poxviruses seen in these regions. number of those being in the Pacific, East North Central While this sample may not be representative of the and South Atlantic region. Four respondents reported see- country’s infectious disease physicians, it likely encom- ing a case of sealpox (one respondent in Providence, Hali- passes the best informed and the least apt to be dis- fax, and Sacramento and one in an unknown location). No suaded by lack of immediate knowledge. Respondents one reported seeing a case of tanapox or a case of oral were able to suggest appropriate infection control mea- rabies vaccine (ORV) related human infection. sures, such as a combination of standard, contact and This survey provided an opportunity for public health droplet precautions for monkeypox and standard precau- practitioners to gain insight into physician approaches tions for orf, and pursued reasonable reporting mechan- to evaluation, diagnosis and reporting of suspected pox- isms. A large proportion of respondents did, however, virus-associated infections. As a result we were able to indicate that they would institute higher then necessary identify potential knowledge gaps and deficits in the transmission precautions for the orf case scenario. We availability of useful resources to facilitate accurate case found that respondents were not necessarily aware of identification and management. diagnostic tests available or where to find them. How- One of the weaknesses of this study was a low survey ever, the majority of respondents did indicate they would response rate. Several factors could account for this. order diagnostic testing for both case scenarios. Anecdotal suggestions are that many potential respon- The public health community can play a greater role ders felt they were being “tested” rather than queried, in reinforcing messages to health practitioners to
Figure 3 Suspected poxvirus infections (omitting molluscum contagiosum) reported by region. Each block in the bar charts represents one respondent. To permit more accurate comparisons of inter-regional variations, graduated green circles show the total number of respondents separated into three classes. Hughes et al. BMC Research Notes 2010, 3:46 Page 6 of 6 http://www.biomedcentral.com/1756-0500/3/46
address appropriate infection control procedures when The findings and conclusions in this report are those of the authors and do dealing with suspected cases of poxvirus infection. They not necessarily reflect the views of the CDC. can also play an active role in disseminating information Author details about new diagnostic tests (such as PCR and serologic 1Centers for Disease Control and Prevention, National Center for Zoonotic, Vector-borne, and Enteric Diseases, Division of Viral and Rickettsial Diseases, tests). In response to findings from the survey, EIN 1600 Clifton Road, NE Atlanta, Georgia, 30333, USA. 2Centers for Disease members were provided a survey report summarizing Control and Prevention, Epidemic Intelligence Service, Office of Workforce CDC recommendations for the various scenarios [Addi- and Career Development, 1600 Clifton Road, NE Atlanta, Georgia, 30333, USA. 3University of Iowa Carver College of Medicine, Department of Internal tional file 2]. We also produced a fact sheet containing Medicine, Iowa City, Iowa, 52242, USA. information pertinent to which diagnostic tests are cur- rently available at CDC and elsewhere for etiologic Authors’ contributions CH, EL, MG and ID conceived of the survey, participated in its design and determination of poxvirus-associated infection [Addi- content, and assisted in the interpretation of results. CH did the statistical tional file 3]. The findings from this survey will also analysis, wrote the manuscript, and created the tables. CH and RL created help guide us in the redesign of CDC’s poxvirus website. the manuscript figures. SB and PP coordinated the survey design, distribution and collection of results. All authors reviewed, revised and We will make infection control practices, diagnostic approved the final manuscript. capabilities, and reporting mechanisms more readily available on the website for physicians. Competing interests Poxviruses occur across the U.S. and around the The authors declare that they have no competing interests. world. Infectious disease physicians may not be the first Received: 5 February 2010 clinicians to see patients with suspected poxvirus infec- Accepted: 25 February 2010 Published: 25 February 2010 tions, but many will be asked to provide expert advice References and consultation. Infectious disease physicians should be 1. Steinhart B: Orf in humans: dramatic but benign. CJEM 2005, 7:417-419. provided with the necessary tools to make well-informed 2. Laboratory-acquired vaccinia exposures and infections–United States, decisions regarding suspected cases of poxvirus infec- 2005-2007. MMWR Morb Mortal Wkly Rep 2008, 57:401-404. 3. Vulvar vaccinia infection after sexual contact with a military smallpox tions. This survey identified key areas in which public vaccinee–Alaska, 2006. MMWR Morb Mortal Wkly Rep 2007, 56:417-419. health practitioners can better serve physicians by focus- 4. Vora S, Damon I, Fulginiti V, Weber SG, Kahana M, Stein SL, Gerber SI, ing on education. These key areas include infection con- Garcia-Houchins S, Lederman E, Hruby D, et al: Severe eczema vaccinatum in a household contact of a smallpox vaccinee. Clin Infect Dis 2008, trol practices and knowledge of various diagnostic tests 46:1555-1561. available for poxviruses. Similar knowledge gaps exist 5. McGuill MW, Kreindel SM, DeMaria A Jr, Robbins AH, Rowell S, Hanlon CA, for other relatively rare diseases that could become Rupprecht CE: Human contact with bait containing vaccine for control of rabies in wildlife. J Am Vet Med Assoc 1998, 213:1413-1417. more common in an outbreak setting. This type of sur- 6. Rupprecht CE, Blass L, Smith K, Orciari LA, Niezgoda M, Whitfield SG, vey could also be helpful in bringing attention to those Gibbons RV, Guerra M, Hanlon CA: Human infection due to recombinant gaps and providing innovative ways to keep physicians vaccinia-rabies glycoprotein virus. N Engl J Med 2001, 345:582-586. 7. Human vaccinia infection after contact with a raccoon rabies vaccine better informed. bait - Pennsylvania, 2009. MMWR Morb Mortal Wkly Rep 2009, 58:1204-1207. Additional file 1: Poxvirus EIN survey. A copy of the poxvirus survey 8. Reed KD, Melski JW, Graham MB, Regnery RL, Sotir MJ, Wegner MV, sent to EIN members Kazmierczak JJ, Stratman EJ, Li Y, Fairley JA, et al: The detection of Click here for file monkeypox in humans in the Western Hemisphere. N Engl J Med 2004, [ http://www.biomedcentral.com/content/supplementary/1756-0500-3-46- 350:342-350. S1.DOC ] 9. Reynolds MG, Yorita KL, Kuehnert MJ, Davidson WB, Huhn GD, Holman RC, Damon IK: Clinical manifestations of human monkeypox influenced by Additional file 2: Final survey report. EIN members were provided route of infection. J Infect Dis 2006, 194:773-780. with this survey report summarizing CDC recommendations for the 10. Croitoru AG, Birge MB, Rudikoff D, Tan MH, Phelps RG: Tanapox virus various scenarios infection. Skinmed 2002, 1:156-157. Click here for file 11. Dhar AD, Werchniak AE, Li Y, Brennick JB, Goldsmith CS, Kline R, Damon I, [ http://www.biomedcentral.com/content/supplementary/1756-0500-3-46- Klaus SN: Tanapox infection in a college student. N Engl J Med 2004, S2.DOC ] 350:361-366. Additional file 3: Poxvirus diagnostic fact sheet. Fact sheet containing 12. The emerging infections network: a new venture for the Infectious information pertinent to which diagnostic tests are currently available at Diseases Society of America. Executive Committee of the Infectious CDC and elsewhere for etiologic determination of poxvirus-associated Diseases Society of America Emerging Infections Network. Clin Infect Dis infection 1997, 25:34-36. Click here for file [ http://www.biomedcentral.com/content/supplementary/1756-0500-3-46- doi:10.1186/1756-0500-3-46 S3.DOC ] Cite this article as: Hughes et al.: Clinical experience, infection control practices and diagnostic algorithms for poxvirus infections - an Emerging Infections Network survey. BMC Research Notes 2010 3:46.
Acknowledgements The authors would like to thank Cynthia Goldsmith, Dr. Christopher Paddock, Whitni Davidson, Zach Braden, Dr. Victoria Olson and Scott Smith for help with the poxvirus diagnostic handout.